InsureNER — Insurance Named Entity Recognition

Created by Bytical AI — AI agents that run insurance operations.

Model Description

InsureNER is a domain-specific Named Entity Recognition model for the UK insurance industry. Built on ModernBERT-base, it recognizes 13 insurance-specific entity types using BIO tagging (26 tags + O = 27 total labels).

Entity Types (13)

Entity	Description	Example
CLAIM_NUMBER	Insurance claim reference	CLM-2024-001234
DATE	Dates in insurance context	15 March 2026
INSURER	Insurance company name	Aviva, AXA, Zurich
LOB	Line of Business	Motor, Property, Liability
MGA	Managing General Agent	Covéa, eSure
MONEY	Monetary amounts	£45,000, $1.2M
ORG	Organisation name	FCA, Lloyd's of London
PERIL	Insurance peril/risk	Flood, Fire, Theft
PERSON	Person name	John Smith
POLICY_NUMBER	Policy reference	POL-UK-2024-56789
POSTCODE	UK postcode	SW1A 1AA, EC2M 7PP
REGULATION	Regulatory reference	Consumer Duty, Solvency II
SYNDICATE	Lloyd's syndicate	Syndicate 2623
VEHICLE	Vehicle description	2023 BMW 320d

Training Details

Parameter	Value
Base Model	answerdotai/ModernBERT-base
Training Samples	8,000 synthetic NER-annotated insurance texts
Epochs	8
Label Schema	BIO (27 labels)
GPU	NVIDIA Tesla T4 16GB

Evaluation Results

Metric	Score
F1	1.0
Precision	1.0
Recall	1.0
Eval Loss	4.80e-05
Eval Samples/sec	68.72

How to Use

from transformers import AutoModelForTokenClassification, AutoTokenizer, pipeline

model = AutoModelForTokenClassification.from_pretrained("piyushptiwari/InsureNER")
tokenizer = AutoTokenizer.from_pretrained("piyushptiwari/InsureNER")

ner_pipeline = pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy="simple")

text = "Aviva policy POL-UK-2024-56789 covers John Smith at SW1A 1AA for motor insurance. Claim CLM-2024-001234 was filed on 15 March 2026 for £45,000."
entities = ner_pipeline(text)

for ent in entities:
    print(f"  {ent['entity_group']:20s} {ent['word']:30s} (score: {ent['score']:.3f})")

Part of the INSUREOS Model Suite

This model is part of the INSUREOS — a complete AI/ML suite for insurance operations built by Bytical AI:

Model	Task	Metric
InsureLLM-4B	Insurance domain LLM	ROUGE-1: 0.384
InsureDocClassifier	12-class document classification	F1: 1.0
InsureNER (this model)	13-entity Named Entity Recognition	F1: 1.0
InsureFraudNet	Fraud detection (Motor/Property/Liability)	AUC-ROC: 1.0
InsurePricing	Insurance pricing (GLM + EBM)	MAE: £11,132

Citation

@misc{bytical2026insurener,
  title={InsureNER: Insurance Named Entity Recognition with ModernBERT},
  author={Bytical AI},
  year={2026},
  url={https://huggingface.co/piyushptiwari/InsureNER}
}

About Bytical AI

Bytical builds AI agents that run insurance operations — claims automation, underwriting intelligence, digital sales, and core system modernization for insurers across the UK and Europe. Microsoft AI Partner | NVIDIA | Salesforce.